[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: dose-response
Could anyone tell me what is the best way to test a dose response relationship
between a binary outcome measure and a four categories exposure. Is there any
method to test dose-response relationship following a binomial regression?
What is best depends....
To help yourself, start asking:
findit dose response
It does not give much, but then try:
findit trend test
Among the hits are this FAQ, which you should read.
FAQ . . . . . . . . . . . . . . A comparison of different tests for trend
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . W. Sribney
3/96 Does Stata provide a test for trend?
Apart from what is mentioned in that FAQ, you can use -tabodds- (even for
stratified analyses; the above FAQ did not mention that). Also, a nonparametric
test like -kwallis- may be useful.
In a regression analysis with binary outcome, e.g. -logistic- and -binreg- you
may include your four-category exposure variable "as is".
Finally, you might want to study Roger Newson's -somersd- package. Find and
install it by -findit somersd-.
But what is best depends...
Hope this helps
Institut for Folkesundhed, Afdeling for Epidemiologi
(Institute of Public Health, Department of Epidemiology)
Vennelyst Boulevard 6
DK-8000 Aarhus C, Denmark
Phone, work: +45 8942 6090
Phone, home: +45 8693 7796
Fax: +45 8613 1580
* For searches and help try: